1 /*
2 * "streamable kanji code filter and converter"
3 * Copyright (c) 1998-2002 HappySize, Inc. All rights reserved.
4 *
5 * LICENSE NOTICES
6 *
7 * This file is part of "streamable kanji code filter and converter",
8 * which is distributed under the terms of GNU Lesser General Public
9 * License (version 2) as published by the Free Software Foundation.
10 *
11 * This software is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU Lesser General Public License for more details.
15 *
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with "streamable kanji code filter and converter";
18 * if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19 * Suite 330, Boston, MA 02111-1307 USA
20 *
21 * The author of this part: Haluk AKIN <halukakin@gmail.com>
22 *
23 */
24 /*
25 * The source code included in this files was separated from mbfilter_ru.c
26 * by moriyoshi koizumi <moriyoshi@php.net> on 4 dec 2002.
27 *
28 */
29
30 #include "mbfilter.h"
31 #include "mbfilter_cp1254.h"
32 #include "unicode_table_cp1254.h"
33
34 static int mbfl_filt_ident_cp1254(int c, mbfl_identify_filter *filter);
35
36 static const char *mbfl_encoding_cp1254_aliases[] = {"CP1254", "CP-1254", "WINDOWS-1254", NULL};
37
38 const mbfl_encoding mbfl_encoding_cp1254 = {
39 mbfl_no_encoding_cp1254,
40 "Windows-1254",
41 "Windows-1254",
42 (const char *(*)[])&mbfl_encoding_cp1254_aliases,
43 NULL,
44 MBFL_ENCTYPE_SBCS,
45 &vtbl_cp1254_wchar,
46 &vtbl_wchar_cp1254
47 };
48
49 const struct mbfl_identify_vtbl vtbl_identify_cp1254 = {
50 mbfl_no_encoding_cp1254,
51 mbfl_filt_ident_common_ctor,
52 mbfl_filt_ident_cp1254
53 };
54
55 const struct mbfl_convert_vtbl vtbl_cp1254_wchar = {
56 mbfl_no_encoding_cp1254,
57 mbfl_no_encoding_wchar,
58 mbfl_filt_conv_common_ctor,
59 NULL,
60 mbfl_filt_conv_cp1254_wchar,
61 mbfl_filt_conv_common_flush,
62 NULL,
63 };
64
65 const struct mbfl_convert_vtbl vtbl_wchar_cp1254 = {
66 mbfl_no_encoding_wchar,
67 mbfl_no_encoding_cp1254,
68 mbfl_filt_conv_common_ctor,
69 NULL,
70 mbfl_filt_conv_wchar_cp1254,
71 mbfl_filt_conv_common_flush,
72 NULL,
73 };
74
75 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
76
77 /*
78 * wchar => cp1254
79 */
80 int
mbfl_filt_conv_wchar_cp1254(int c,mbfl_convert_filter * filter)81 mbfl_filt_conv_wchar_cp1254(int c, mbfl_convert_filter *filter)
82 {
83 int s, n;
84
85 if (c < 0x80) {
86 s = c;
87 } else {
88 s = -1;
89 n = cp1254_ucs_table_len-1;
90 while (n >= 0) {
91 if (c == cp1254_ucs_table[n] && c != 0xfffe) {
92 s = cp1254_ucs_table_min + n;
93 break;
94 }
95 n--;
96 }
97 if (s <= 0 && (c & ~MBFL_WCSPLANE_MASK) == MBFL_WCSPLANE_CP1254) {
98 s = c & MBFL_WCSPLANE_MASK;
99 }
100 }
101
102 if (s >= 0) {
103 CK((*filter->output_function)(s, filter->data));
104 } else {
105 CK(mbfl_filt_conv_illegal_output(c, filter));
106 }
107
108 return c;
109 }
110
111 /*
112 * cp1254 => wchar
113 */
114 int
mbfl_filt_conv_cp1254_wchar(int c,mbfl_convert_filter * filter)115 mbfl_filt_conv_cp1254_wchar(int c, mbfl_convert_filter *filter)
116 {
117 int s;
118
119 if (c >= 0 && c < cp1254_ucs_table_min) {
120 s = c;
121 } else if (c >= cp1254_ucs_table_min && c < 0x100) {
122 s = cp1254_ucs_table[c - cp1254_ucs_table_min];
123 if (s <= 0) {
124 s = c;
125 s &= MBFL_WCSPLANE_MASK;
126 s |= MBFL_WCSPLANE_CP1254;
127 }
128 } else {
129 s = c;
130 s &= MBFL_WCSGROUP_MASK;
131 s |= MBFL_WCSGROUP_THROUGH;
132 }
133
134 CK((*filter->output_function)(s, filter->data));
135
136 return c;
137 }
138
139 /* We only distinguish the MS extensions to ISO-8859-1.
140 * Actually, this is pretty much a NO-OP, since the identification
141 * system doesn't allow us to discriminate between a positive match,
142 * a possible match and a definite non-match.
143 * The problem here is that cp1254 looks like SJIS for certain chars.
144 * */
mbfl_filt_ident_cp1254(int c,mbfl_identify_filter * filter)145 static int mbfl_filt_ident_cp1254(int c, mbfl_identify_filter *filter)
146 {
147 if (c >= 0x80 && c < 0xff)
148 filter->flag = 0;
149 else
150 filter->flag = 1; /* not it */
151 return c;
152 }
153